cognition in movement disorders: where can we hope to be in ten years?

Cognition in Movement Disorders: Where Can WeHope to be in Ten Years?

David Burn, MD,1* Daniel Weintraub, MD,2 Bernard Ravina, MD,3 and Irene Litvan, MD4

1Institute for Aging and Health, Newcastle University, Newcastle upon Tyne, UK2Perelman School of Medicine at the University of Pennsylvania, Philadelphia, Pennsylvania, USA

3Neurology Clinical Development, Biogen Idec, Cambridge, Massachusetts, USA4Movement Disorder Center, Department of Neurosciences, University of California, San Diego, California, USA

ABSTRACT: Cognitive impairment and dementiaassociated with movement disorders represent a majormanagement challenge and area of unmet need. This arti-cle has focused upon Parkinson’s disease as an exemplarcondition, but many of the roadblocks and efforts to over-come these are applicable, in a general sense, to otherdisorders. Short of a “penicillin moment”—a chance dis-covery or piece of unintended good fortune—progress islikely to be incremental. Cognitive therapies may end up

being multiple and possibly multimodal, parallel with thecancer therapy field. Ultimately, benefit for one conditionmay extend to others as commonality in protein aggrega-tion, synergistic pathological effects between proteins,and pathological spread emerges. VC 2014 InternationalParkinson and Movement Disorder Society

Key Words: Parkinson’s disease; dementia; cogni-tive testing; biomarkers; therapeutics

Despite a greater knowledge of the pathophysiologyunderpinning several movement disorders, this has notyet translated into major improvements in our clinicalmanagement of the devastating cognitive problems asso-ciated with these conditions. Cholinesterase inhibitorsprobably represent the most notable therapeutic stepforward in the field over the past decade, based uponneurotransmitter augmentation, a conceptually simpleprinciple. Yet, the failure of such agents to improve cog-nitive dysfunction in four-repeat tauopathies such asPSP, where there is profound cholinergic deficiency (andeven, in some cases, to have counterproductive effects),highlights that even the benefits of manipulating neuro-transmitter function cannot be generalized.

Unfortunately, more fundamental approaches of dis-ease modification have, as yet, failed to yield benefit.There are, undoubtedly, several reasons behind this fail-

ure, including lack of understanding of disease patho-physiology and/or inability to identify tractable targets,inadequate animal models, disease heterogeneity, and anabsence of robust biomarkers for disease progression.

In 10 years, a breakthrough could come from discover-ing a tractable target and drug development that tacklesbasic disease progression; in other words, both motorand cognitive deficits could both be delayed or favorablymodified. Alternatively, identifying people at high risk ofdementia, for example, in association with Parkinson’sdisease (PD), might lead to the use of transferrable tech-nologies or drugs from other disease areas, such as Alz-heimer’s disease (AD), that may also have efficacy in PD.

This article adopts a “blue skies” approach to wherewe might be in 10 years, addressing how we mightassess the patient, trial design, new drugs in the pipe-line, and alternative nonpharmacological approaches.We will focus upon PD as an exemplar condition.

Where Can We Hope to be inAssessment of the Patient’s

Cognition?

Prioritizing Serial, Long-Term CognitiveTesting for Clinical Care and Research

Ideally, results from neuropsychological testingadministered before PD onset would be available for

------------------------------------------------------------*Correspondence to: Dr. David J. Burn, Clinical Aging Research Unit,Newcastle University, Campus for Aging and Vitality, Newcastle uponTyne, NE4 5PL, UK; [email protected]

Grant funding: NIA 5R01AG024040 (to I.L.).

Relevant conflicts of interest/financial disclosures: Nothing to report.Full financial disclosures and author roles may be found in the online ver-sion of this article.

Received: 17 December 2013; Revised: 10 January 2014;Accepted: 27 January 2014

Published online in Wiley Online Library(wileyonlinelibrary.com). DOI: 10.1002/mds.25850

R E V I E W

704 Movement Disorders, Vol. 29, No. 5, 2014

comparison, but this is rarely the case. When available,recent and previous performance can be comparedinformally or by using a reliable change index, whichtakes into account measurement error and practiceeffects, when longitudinal group-level data are avail-able. Given demographic trends and the current empha-sis on identifying neurodegenerative diseases at theearliest stage possible, a case can be made that, movingforward, everyone should receive brief, global neuro-psychological testing on a regular basis starting in mid-life. Such testing could be done in the context of rou-tine primary care, need only occur every 3 to 5 years,and could utilize tests that only take 10 minutes tocomplete (e.g., the Montreal Cognitive Assessment;MoCA). If this is not realistic, then the next-best optionis to have all PD patients undergo some form of cogni-tive evaluation annually, starting at the time of diagno-sis. Given that 80% to 85% of PD patients havenormal cognition at diagnosis, this will allow a reason-able cognitive baseline to be established for futurereference.

Emergence of Performance-Based Assessmentof Functional Abilities That Are Sensitive and

Specific to Cognitive Performance

Another criterion for a diagnosis of cognitive disor-der is assessment of cognition-related functional abil-ities. Currently, there is no consensus on whatconstitutes significant functional impairment. Compli-cating the issue in PD is the impact of motor deficitson task completion, whether or not to consider“slowness,” but otherwise normal performance a cog-nitive deficit, and the broad age range for PD (com-pared with AD) that contributes to significantinterindividual variability in everyday functionaldemands that makes comparisons across individualsproblematic.

An additional issue relates to how best to assessfunction as it relates to cognitive abilities. In clinicalpractice, this is determined on the basis of an unstruc-tured encounter. However, for clinical research, it isimportant to have validated measures to assess func-tion not only to increase accuracy of assigned diagno-ses, but also to demonstrate worsening (i.e., withdisease progression) or improvement (i.e., with cogni-tive enhancing treatment) over time.

Currently, function in clinical research is assessedthrough an interview (with the patient, an informedother, or both) by inquiring about performance ofhigher-level instrumental activities of daily living.Instruments such as the Alzheimer’s Disease Co-operative Study/Activities of Daily Living Inventoryscale have been used, but PD-related motor impair-ment potentially biases the rating of some activitiesand the different cognitive profile of PD limits the util-ity of instruments designed for other populations.

Recently, PD cognition-specific functional instrumentshave been developed and validated, including the PD-Cognitive Functional Rating Scale1 and the Penn DailyActivities Questionnaire.2

Ultimately, because of the subjectivity and impreci-sion of questionnaires, further development and test-ing of performance-based measures for cognition inPD is needed. Examples of measures developed in thegeneral population, but applied in PD to assess generalor specific abilities, include the Everyday CognitiveBattery (which assesses medication use, finances, nutri-tion, and treatment choices),3 the Hopkins MedicationSchedule,4 the Capacity to Consent to Treatment,5 theUniversity of California San Diego Performance-basedSkills Assessment,6 and the MacArthur CompetenceAssessment Tool for Clinical Research.7 In addition,driving safety is correlated with executive and visuo-spatial abilities in PD,8 and simulated driving tests foruse in clinical research are now available. In thefuture, having a battery of validated performance-based measures, including computerized testing, wouldadvance the field of PD cognition research and couldimprove clinical practice. Such instruments can also beused set a minimum clinically important difference(MCID), which is increasingly used in clinical trials todetermine the clinical significance of treatment effects.It is also likely that new APP-based tools for mobilephones and tablets will become more frequently usedin the assessment and monitoring of cognitive func-tion, whereas body-worn sensors may give objectivedata regarding daily activity patterns and overall levelsof function. Such data may now be analyzed morerapidly and meaningfully with advanced computerizedalgorithms.

Consensus on Preferred NeuropsychologicalTests: Is it Desirable and Achievable?

Traditional global cognitive instruments developedfor non-PD populations, but often used in PD, havebeen largely replaced by instruments developed orvalidated for use in PD. In addition, a large number ofindividual neuropsychological tests have been used toassess executive, memory, attention, visuospatial, andlanguage abilities specifically.

On the one hand, use of a variety of instrumentsprovides rich detail about the relative cognitivestrengths and weaknesses of PD patients. However,the plethora of tests used also limits progress. Formost existing instruments, major holes regarding per-formance need to be filled with additional research.None, other than the MoCA for screening purposes,has gained wide acceptance in PD. Given the lack ofconsensus regarding current instruments, a case can bemade to assemble a neuropsychological battery for usein all aspects of PD cognitive research. A “wish list”for such a battery would include:

C O G N I T I O N I N M O V E M E N T D I S O R D E R S

Movement Disorders, Vol. 29, No. 5, 2014 705

� Existing neuropsychological tests, sensitive toearly cognitive decline in PD� Tests with normative data to allow calculation of

normed test, domain, and battery scores� Cover comparably the five primary cognitive

domains (attention-working memory, memory,executive functioning, visuospatial skills, andlanguage)� At least two test scores per domain to allow

application of level II criteria for mild cognitiveimpairment associated with PD� Assess reliability and validity for use in PD

patients with normal cognition, mild cognitiveimpairment, and mild dementia� Sensitive to worsening in cognitive abilities over

time� Sensitive to improvements with cognitive enhanc-

ing treatment� To limit motor confounds (i.e., limit tests that

require motor skills)� To limit speed confounds (i.e., limit timed tests

for primary cognitive domains)� Both paper- and computer-based versions of the

instrument� Alternate versions for serial evaluation in longitu-

dinal studies� Appropriate for translation from English into

multiple languages� To administer to a large number of healthy

elderly to generate age, education, and gendernorms for this battery specifically� Establish an MCID� Demonstrate a correlation with measures of cog-

nitive functioning� Demonstrate a correlation with PD cognitive

biomarkers

Use of Technology

Although concerns have been raised about reliabilityand validity, which still need to be addressed withadditional research, there are many potential advan-tages to computerized testing. Because many clinicalcenters do not have ready access to neuropsycholo-gists, the PD clinical population is currently under-served. The nature of the computerized instrumentsallows administration by health care personnel otherthan neuropsychologists. Computerized testing offers aclear advantage in terms of standardized administra-tion procedures, precise timing of stimulus presenta-tion and response time, and automated scoring andresults generation. In comparison with traditional neu-ropsychological assessment instruments, computerizedtests may also represent a potential cost savings notonly with regard to materials and supplies, but also inthe time required of the test administrator. In

addition, testing can occur in variety of environments,and alternate versions can be easily developed.

Will Trial Design Have Changed?

Will Better Animal Models Inform Trial Designand Size?

Therapeutic trials using better animal models couldhelp accelerate identification of effective interventionsand could provide information that would be morerealistic for designing clinical trials in humans, includ-ing estimating sample size and trial duration. At mini-mum, using better animal models could help advanceour understanding of the underlying neural mecha-nisms influencing cognitive decline in patients withmovement disorders. Currently, animal trials of move-ment disorders are usually conducted in rodent modelsusing suboptimal methodologies, including small sam-ple size and nonstandardized tests. There are signifi-cant limitations in testing complex human cognitivefunctions using current animal models, although thereare some human tests that could have an comparableequivalents in animal models.9 For example, humanexecutive function set shifting and inhibition of behav-ior tasks could relate to delayed alternation tests andstop task tests in animals, but planning and multitask-ing have no analogous animal model test. Moreover,at present, most therapeutic preclinical studies con-ducted in animal models test therapeutic agents at pre-clinical disease stages, rather than when diseases areclinically manifest or even at advanced stages as theyare evaluated in humans. Thus, to date, results of neu-rotransmitter replacement and biological therapies,shown to be very successful in animal models, havenot been of demonstrable benefit in humans. After somany therapeutic failures, one would hope that in 10years industry and academia would be conductingstudies in nonhuman primates (e.g., macaques, mon-keys, and marmosets), in which cognition could bebetter tested and in which pharmacokinetics, pharma-codynamics, and cerebrospinal fluid (CSF) brain pene-tration and drug brain levels may be similar to thosein humans.10,11

Future animal model trials will hopefully apply thesame meticulous methodology used in human clinicaltrials, including sample-size estimation, standardiza-tion of tests, blinded evaluations, and administrationof therapies at clinical disease stages. These studieswill be more expensive, but higher costs up front mayallow cost cutting at later stages of development andincrease chances of success. However, even if animalmodels and methodology improve, we should beaware that studies in such models will not capture theheterogeneity observed in humans who also exhibit avariety of comorbidities as well as lifestyle, genetics,and environmental exposures.

B U R N E T A L


Toward Improved Biomarkers

Well-defined populations are critical for clinical tri-als, but there are no disease-specific biomarkers forthe diagnosis of nongenetic movement disorders. Toevaluate targeted interventions and prevention thera-pies designed to delay the cognitive decline observedin these disorders, we need biomarkers that couldincrease diagnostic accuracy, predict cognitive decline,and accurately track progression of this decline.

Diagnostic biomarkers that quantify levels of CSFbeta-amyloid, tau and phosphorylated tau, and PETimaging tracers that have high affinity for beta-amyloid are useful in the diagnosis of AD in vivo, forwhich they have been recently incorporated toimprove its diagnostic criteria.12 However, whereasquantification of CSF alpha-synuclein (a-Syn) levels isa promising biomarker for the diagnosis of PD,13

higher specificity and sensitivity are needed to allowits adoption to refine extant diagnostic criteria. Fur-thermore, CSF tau levels are not helpful for diagnosingtauopathies presenting with parkinsonism, such as PSPand corticobasal degeneration. However, the recentdevelopment of two PET imaging tracers targetingpaired helical filament-tau, phenyl/pyridinyl-buta-dienyl-benzothiazoles/benzothiazoliums (PBBs) and[(18)F]T807, that have high affinity in vitro, in animalmodels and in patients with AD are encouraging andmay help increase the accuracy in diagnosing thesedisorders.14-16 Validation of these biomarkers in AD isin progress. Evidence that these biomarkers will behelpful to diagnose patients with other tauopathies isnot yet available, although one report describesincreased [11C]PBB signal in areas typically affected ofa patient with a corticobasal syndrome.

This new development increases the likelihood offinding tracers that could label a-Syn because thisshows us that imaging radiotracers can feasibly labelintracellular protein inclusions such as those found inPD.17 Improved biomarkers not only will increasediagnostic accuracy, but will also allow the diagnosisof patients at earlier disease stages, thereby improvingthe chances of finding successful biological therapies.PET tau and a-Syn biomarkers may also serve as sur-rogate markers of disease progression. It is unlikelythat PET ligands will become inexpensive and accessi-ble world-wide, but they could pave the way for theidentification of other less-costly, more-ubiquitousimaging approaches (e.g. single-photon emission CT),as well as less-invasive blood or urine biomarkers thatcould allow an early diagnosis of movement disorders.

Moreover, in the coming years, we may also be ableto use genetics to design therapeutic trials that willtarget specific populations more likely to respond totherapy or with similar disease progression, which, inturn, will increase the likelihood of finding successfultherapies. For example, if apolipoprotein E4 (ApoE4),

the H1/H1 haplotype, or the a-Syn gene predispose tothe conversion of PD patients with mild cognitiveimpairment to dementia, stratifying or selectingpatient samples with these genes will decrease thesample size needed and may more readily identify suc-cessful therapies.

Adaptive Design/Fast Tracking

In an attempt to increase the efficiency and likeli-hood of success, there is a growing interest amongdrug developers in using adaptive design approaches,defined as an a priori plan to modify aspects of astudy based on interim data analyses performed atpreplanned time points.18 This approach could lead tofundamental “in flight” changes that include totalsample size, dose level, concomitant treatments,changes in primary or secondary outcomes, and studyduration (e.g., early termination). There are concerns,however, that this approach could introduce bias andincrease the possibility of type 1 errors (incorrect con-clusions). Adjustment of the sample size during thecourse of a clinical trial has become a popular strategylately.19

Drugs in the Pipeline:Hope or hype?

Therapeutic efforts in PD will increasingly focus oncognitive decline over the next decade. This will bedriven by the high unmet need and growing under-standing of the pathophysiology and clinical course ofcognitive decline. It appears that cognitive impairmentis multifactorial,20 and progress will therefore beincremental. The biology of cognitive impairment andan appreciation of overlapping features with AD mayallow the field to leverage biomarkers and interven-tions from the AD field. This could accelerate the paceat which new interventions are tested.

Clinical trials of agents for cognitive impairmentmay be aimed at enhancing cognition or slowingdecline. Pharmacological or biological interventionsmay focus on a range of targets, from fundamentaldisease processes to resulting neurotransmitter deficits,such as acetylcholinesterase (AChE) inhibitors. Near-term successes are likely to build on the modest, butreal, progress in alleviating neurotransmitter deficits.Though treatment effects are small, and similar inmagnitude to AD, successful AChE inhibitor trials val-idate the concept of enhancing cholingergic transmis-sion as a therapeutic approach.21-23 They also suggestthat trials in AD may be predictive for PDD, at leastfor certain mechanisms.

Selective cholinergic agonists may provide additionalbenefit beyond the cholinesterase inhibitors. EVP-6124, a selective nicotinic alpha 7 coagonist, showedsignificant improvement on an abbreviated version of



the Alzheimer’s Disease Assessment Scale, as well asthe clinical dementia rating scale sum of boxes. As acoagonist, the agent works in combination with ace-tylcholine (ACh), making it possible for smalleramounts of naturally occurring ACh to be effective inactivating the alpha 7 receptor and possibly improvingthe tolerability, as compared with previous nicotinicagonists.24 There have been other promising alpha 7clinical trials in the AD population and in subjectswith cognitive impairment in the setting of schizophre-nia. Partial agonists of the alpha 4/beta 2 receptor andother mechanisms of targeting cholinergic pathwayshave also shown some promise. If the effects are con-firmed, the prominent cholinergic deficits would makePD with dementia (PDD) a natural next step for clini-cal trials with these agents.

Other neurotransmitter-based approaches may be via-ble in PD-related cognitive impairment. Agents thatinhibit the excitotoxicity of glutamate have been chal-lenging to develop. There is inconsistent evidence thatthe N-methyl D-aspartate receptor antagonist, meman-tine, has any benefit in PDD.25,26 The effect of dopami-nergic replacement strategies on cognition in PD remainsuncertain. Exogenous dopamine replacement has variableeffects, which may depend on the selectivity for dopa-mine receptors, particular cognitive domains assessed,and the stage of disease. However, there appears to be aclear relationship between cognition and dopamine defi-ciency, as measured by dopamine transporter imag-ing,27,28 and dopamine metabolism, as assessed bycatechol-O-methyltransferase polymorphisms.29 Thesefindings are consistent with the role of dopamine in sup-porting cognitive function underpinned by frontostriatalconnections. Therefore, any strategy that protects orrestores nigral dopaminergic neurons, and allows formore physiologic dopaminergic transmission than isobtained with exogenous dopamine administration, couldpreserve or enhance specific cognitive domains.

Treatments aimed at the underlying etiology may

slow, delay, or even prevent cognitive decline. There is

a strong case to be made for a-Syn-related pathologyin the development of cognitive impairment. Human

genetics,30 clinical pathological correlations,31 and

experimental models32 all support the role of a-Syn

pathology as a driver of PDD. Therefore, approaches

to reducing aggregation or limiting the spread of Lewy

pathology may slow cognitive decline. These are early

days for such approaches and little is known aboutthe right species or form (monomer, oligomers, or

aggregates) of a-Syn to target, the safety of altering a-

Syn levels, or the extent of change that is necessary

for clinical benefit. However, experimental models

suggest that active or passive immunotherapy

approaches can reduce a-Syn levels, as measured bybiochemical assays or immunohistochemistry, and are

associated with improvement in cognitive behaviors.33

There are important challenges for any programs

targeting a-Syn pathology. They must first show safetyand then biological activity. This will be difficult,

given that there is currently no a-Syn imaging marker.

Total CSF a-Syn may be useful for this purpose. How-

ever, this has not performed particularly well as a

diagnostic or prognostic marker. Several unknowns

may make this biomarker difficult to use and interpret,including the uncertain relationship of extra- and

intracellular a-Syn levels. The first a-Syn immunother-

apy program, an active vaccination, is in early clinical

testing and this first wave of clinical studies will help

to determine the overall safety of such programs.Other approaches to disease modification may be

particularly important for slowing the rate of cognitivedecline. Mutations in glucocerebrosidase (GCase) arethe most common genetic risk factor for PD.30 Areciprocal relationship has been proposed betweenreduced GCase activity and a-Syn aggregation.34

GCase carriers with PD appear to have more cognitiveimpairment, and GCase mutations are associated withDLB.35 This suggests that approaches to enhanceenzyme function, from small-molecule chaperones toenzyme replacement with gene therapy, could slowdisease progression and cognitive decline.

b-amyloid and tau may be important targets forcognitive impairment in PD, in addition to their rolein AD. The relative contribution of AD-related pathol-ogy has been controversial: Genetics, human bio-marker studies, and postmortem findings need to beconsidered in determining whether these targets shouldbe pursued. The association of ApoE4 with PDD hasbeen inconsistent, but this is likely a result of meth-odological issues. A recent study shows a consistentassociation with dementia in synucleinopathies.36 Sev-eral postmortem studies have shown a correlationbetween PDD and b-amyloid pathology. Recent datasuggest that both b-amyloid plaques and tau-containing neurofibrillary tangles are contributors toPDD; up to 50% of postmortem brains from patientswith PDD met criteria for a secondary diagnosis ofAD.20 Studies assessing CSF amyloid b(1-42) haveshown faster rates of cognitive decline in those withlower levels,37 as well as a potential association withthe postural instability gait disorder phenotype,38

which itself is associated with a more aggressivecourse and faster cognitive decline. Imaging studiesusing primarily Pittsburgh compound B (PIB) and PETpaint an inconsistent picture. It certainly appears thatfibrillar b-amyloid, which is labeled by most tracers, isgenerally lower than it is in AD or DLB.39 However,PD subjects with higher-than-average binding appearto have faster rates of cognitive progression.40

There are several anti-b-amyloid immunotherapiesin clinical development for AD. The right species orform to target remains unclear. The tau biology field

B U R N E T A L


is at an earlier stage of development than b-amyloid.However, a similar case can be made for the role oftau based on genetics and pathology. The questionwill remain of how important these AD-related path-ologies are for PDD, and the only way to answer thiswill be with natural history studies using relevantimaging and CSF biomarkers. Because of the moreevolved imaging tools, b-amyloid and tau may bemore-testable protein aggregates to pursue in the nearfuture, rather than a-Syn-based therapies.

Although the progression of cognitive dysfunction isrelatively slow, cognitive decline may be a valuableparadigm for testing a range of putative disease-modifying agents. There is a general need for newclinical trial paradigms in PD. The traditional designhas focused on de novo subjects and motor progres-sion. This is becoming more difficult with earlier useof dopaminergic therapy, and the de novo period isthe one in which our current therapies are most effica-cious. There is considerable interest in conversionfrom premotor PD to manifest motor PD as a clinicaltrial paradigm. Low predictive value of premotor fea-tures and long timelines for conversion are likely tomake this a difficult challenge for drug developers.This approach may only be feasible in select, geneti-cally at-risk cohorts, such as leucine-rich repeat kinase2 carriers or GCase mutation heterozygotes, in whomdisease onset starts at a specific age. In contrast, cogni-tive decline is a common feature of PD that progressesrelentlessly and is relatively unaffected by existingtherapies, making it a good marker for disease pro-gression. Diagnostic criteria for mild cognitive impair-ment, PDD, and natural history studies that focus onthe course of cognitive decline and specific cognitivedomains will pave the way for using cognitive declineas a key measure of disease progression in diseasemodification clinical trials with a range of relevantagents.

Alternative Approaches

The potential benefit of exercise and “brain train-ing” on cognitive decline and dementia in PD needs tobe better established. The level of mid-life cardiores-piratory fitness has been associated with lower risk ofdeveloping all-cause dementia in later life.41 In cogni-tively normal older adults, exercise engagement hasbeen associated with reduced PIB binding, as well ashigher cerebrospinal amyloid Ab42 levels,42 whereasgreater levels of physical activity may be associatedwith higher cerebral white matter integrity43 andincreased gray matter volumes. These effects may bemodified, in part, by APO e4 status.44

A recent systematic review studied evidence forefficacy of nonpharmacological therapies in managingcognitive impairment and dementia in PD.45 Cognitive

rehabilitation, physical rehabilitation, exercise, and brainstimulation techniques were considered in this review.The researchers concluded that research in this area iscurrently very limited in both quality and quantity.Despite this, there is some evidence that aerobic exerciseincreases executive function and spatial memory inPD.46,47 The benefits of exercise on cognition in PD maybe mediated by raised levels of brain-derived neurotro-phic or other trophic factors, increased blood flow, wide-spread effects upon the immune system, neurotransmittermodulation (e.g., glutamatergic neurotransmission),increased dendritic spine density, neurogenesis, or combi-nations of these factors.48 Clearly, further work isrequired to better elucidate the mechanism of action ofexercise and the potential size and sustainability of itseffect upon cognitive function in PD.

Very few studies have addressed the potential bene-fits of cognitive training (e.g., regular completion ofSudoku puzzles) upon mental functioning in PD, eitheralone or in combination with exercise training. Mosthave been nonrandomized or recruited small numbersof participants.45 This needs to be addressed in futurestudies. As “proof of principle” that such trainingmay have real-world effects, a recent randomizedstudy showed that cognitive speed of processing train-ing could improve useful field-of-view test (UFOV)performance in people with PD.49 UFOV is a robustpredictor of driving performance in aging and PD.

Other potential, nonpharmacological approachesinclude exploring novel DBS targets. Freund et al.reported on the case of a 71-year-old man with PDD,in whom two electrodes were inserted into the cholin-ergic nucleus basalis of Meynert (nbM), in addition tothe STN.50 Activation of the nBM electrodes resultedin significantly improved attention, concentration,alertness, drive, and spontaneity, which reversed whenthe stimulator system was switched off. In AD, clinicaland preclinical studies have explored the fornix, ento-rhinal cortex, and nbM as potential DBS targets.51

Laxton et al. performed the first DBS study in 6patients with AD, chronically stimulating the fornix/hypothalamus area over 12 months.52 There was evi-dence for reduced rate of decline in Mini–Mental StateExamination scores, as well as increased cerebral cort-ical metabolism in these subjects, with autonomic andcardiovascular side effects only observed at high stim-ulation settings. Further trials are now ongoing in AD,whereas in PDD, the nBM is being actively exploredfurther as a DBS target, given its involvement in mem-ory, attention, arousal, and perception.53 Given therelative preservation of postsynaptic M1 cholinergicreceptors in PDD, compared with AD, it is plausiblethat enhancing ACh release through nBM stimulationmay be associated with significant clinical benefit.

Transcranial magnetic stimulation and transcranialdirect current stimulation (tDCS) are noninvasive



techniques that may improve cognitive function incognitively normal and demented individuals. Theyhave not been fully evaluated in the context of PD,although a small study suggested that active tDCS ofthe left dorsolateral prefrontal cortex could improveworking memory in PD.54 It is conceivable that wear-able neurostimulators, in conjunction with mobileEEG, may be developed to deliver tDCS when EEGsignatures of confusion or hallucinosis are detected.More speculatively, in the future, optogeneticallybased technology may be used to target specific inter-neuronal populations to reduce phenomena such asvisual hallucinations.

References1. Kulisevsky J, Fernandez de Bobadilla R, Pagonabarraga J, et al.

Measuring functional impact of cognitive impairment: validationof the Parkinson’s Disease Cognitive Functional Rating Scale. Par-kinsonism Relat Disord 2013;19:812-817.

2. Weintraub D, Shea J, Rubright J, et al. Psychometric properties ofthe brief Penn Daily Activities Questionnaire (PDAQ) for Parkin-son’s disease. Mov Disord 2013;28:308.

3. Young TL, Granic A, Chen TY, et al. Everyday reasoning abilitiesin persons with Parkinson’s disease. Mov Disord 2010;25:2756-61.

4. Manning K, Clarke C, Lorry A, et al. Medication management andneuropsychological performance in Parkinson’s disease. Clin Neu-ropsychol 2012;26:45-58.

5. Martin RC, Okonkwo OC, Hill J, et al. Medical decision-makingcapacity in cognitively impaired Parkinson’s disease patients with-out dementia. Mov Disord 2008;23:1867-1874.

6. Pirogovsky E, Martinez-Hannon M, Schiehser D, et al. Predictorsof performance-based measures of instrumental activities of dailyliving in nondemented patients with Parkinson’s disease. J ClinExp Neuropsychol 2013;9:926-933.

7. Karlawish J, Cary M, Moelter S, et al. Cognitive impairment and PDpatients’ capacity to consent to research. Neurology 2013;81:801-807.

8. Amick MM, Grace J, Ott B. Visual and cognitive predictors ofdriving safety in Parkinson’s disease patients. Arch Clin Neuropsy-chol 2007;22:957-967.

9. Alexander GE, Ryan L, Bowers D, et al. Characterizing cognitiveaging in humans with links to animal models. Front Aging Neuro-sci 2012;4:21.

10. Hoshi Y, Uchida Y, Tachikawa M, Inoue T, Ohtsuki S, TerasakiT. Quantitative atlas of blood-brain barrier transporters, receptors,and tight junction proteins in rats and common marmoset. J PharmSci 2013;102:3343-3355.

11. Syv€anen S, Lindhe O, Palner M, et al. Species differences in blood-brain barrier transport of three positron emission tomographyradioligands with emphasis on P-glycoprotein transport. DrugMetab Dispos 2009;37:635-643.

12. Jack CR, Jr., Albert MS, Knopman DS, et al. Introduction to therecommendations from the National Institute on Aging-Alzheimer’s Association workgroups on diagnostic guidelines forAlzheimer’s disease. Alzheimers Dement 2011;7:257-262.

13. Wang Y, Shi M, Chung KA, et al. Phosphorylated a-synuclein inParkinson’s disease. Sci Transl Med 2012;4:121ra20.

14. Maruyama M, Shimada H, Suhara T, et al. Imaging of tau pathol-ogy in a tauopathy mouse model and in Alzheimer patients com-pared to normal controls. Neuron 2013;79:1094-1108.

15. Xia CF, Arteaga J, Chen G, et al. [(18)F]T807, a novel tau posi-tron emission tomography imaging agent for Alzheimer’s disease.Alzheimers Dement 2013;9:666-676.

16. Blagbrough IS, Zara C. Animal models for target diseases in genetherapy—using DNA and siRNA delivery strategies. Pharm Res2009;26:1-18.

17. Fodero-Tavoletti MT, Mulligan RS, Okamura N, et al. In vitrocharacterisation of BF227 binding to alpha-synuclein/Lewy bodies.Eur J Pharmacol 2009;617:54-58.

18. Wang SJ, Hung HM, O’Neill R. Paradigms for adaptive statisticalinformation designs: practical experiences and strategies. Stat Med2012;31:3011-3023.

19. G, Shih WJ, Xie T, Lu J. A sample size adjustment procedure for clin-ical trials based on conditional power. Biostatistics 2002;3:277-287.

20. Irwin DJ, White MT, Toledo JB, et al. Neuropathologic substratesof Parkinson disease dementia. Ann Neurol 2012;72:587-598.

21. Ravina B, Putt M, Siderowf A, et al. Donepezil for dementia inParkinson’s disease: a randomised, double blind, placebo con-trolled, crossover study. J Neurol Neurosurg Psychiatry 2005;76:934-939.

22. Emre M, Aarsland D, Albanese A, et al. Rivastigmine for dementiaassociated with Parkinson’s disease. N Engl J Med 2004;351:2509-2518.

23. Dubois B, Tolosa E, Katzenschlager R, et al. Donepezil in Parkin-son’s disease dementia: a randomized, double-blind efficacy andsafety study. Mov Disord 2012;27:1230-1238.

24. Prickaerts J, van Goethem NP, Chesworth R, et al. EVP-6124, anovel and selective a7 nicotinic acetylcholine receptor partial ago-nist, improves memory performance by potentiating the acetylcho-line response of a7 nicotinic acetylcholine receptors.Neuropharmacology 2012;62:1099-1110.

25. Aarsland D, Ballard C, Walker Z, et al. Memantine in patientswith Parkinson’s disease dementia or dementia with Lewy bodies:a double-blind, placebo-controlled, multicentre trial. Lancet Neurol2009;8:613-618.

26. Emre M, Tsolaki M, Bonuccelli U, et al. Memantine for patientswith Parkinson’s disease dementia or dementia with Lewy bodies:a randomised, double-blind, placebo-controlled trial. Lancet Neu-rol 2010;9:969-977.

27. Arnaldi D, Campus C, Ferrara M, et al. What predicts cognitivedecline in de novo Parkinson’s disease? Neurobiol Aging 2012;33:1127.e11-20.

28. Ravina B, Marek K, Eberly S, et al. Dopamine transporter imagingis associated with long-term outcomes in Parkinson’s disease. MovDisord 2012;27:1392-1397.

29. Williams-Gray CH, Evans JR, Goris A, et al. The distinct cognitivesyndromes of Parkinson’s disease: 5 year follow-up of the Cam-PaIGN cohort. Brain 2009;132:2958-2969.

30. Singleton AB, Farrer MJ, Bonifati V. The genetics of Parkinson’sdisease: progress and therapeutic implications. Mov Disord 2013;28:14-23.

31. Halliday GM, McCann H. The progression of pathology in Parkin-son’s disease. Ann NY Acad Sci 2010;1184:188-195.

32. Lim Y, Kehm VM, Lee EB, et al. a-Syn suppression reverses synap-tic and memory defects in a mouse model of dementia with Lewybodies. J Neurosci 2011;31:10076-10087.

33. Masliah E, Rockenstein E, Mante M, et al. Passive immunizationreduces behavioral and neuropathological deficits in an alpha-synuclein transgenic model of Lewy body disease. PLoS One 2011;29:6:e19338.

34. Mazzulli JR, Xu YH, Sun Y, et al. Gaucher disease glucocerebrosi-dase and a-synuclein form a bidirectional pathogenic loop in synu-cleinopathies. Cell 2011;146:37-52.

35. Nalls MA, Duran R, Lopez G, et al. A multicenter study of gluco-cerebrosidase mutations in dementia with Lewy bodies. JAMANeurol 2013;70:727-735.

36. Tsuang D, Leverenz JB, Lopez OL, et al. APOE e4 increases riskfor dementia in pure synucleinopathies. JAMA Neurol 2013;70:223-228.

37. Siderowf A, Xie SX, Hurtig H, et al. CSF amyloid {beta} 1-42 pre-dicts cognitive decline in Parkinson disease. Neurology 2010;75:1055-1061.

38. Kang JH, Irwin DJ, Chen-Plotkin AS, et al. Association of cerebro-spinal fluid b-amyloid 1-42, T-tau, P-tau181, and a-synuclein lev-els with clinical features of drug-naive patients with earlyParkinson disease. JAMA Neurol 2013;70:1277-1287.

39. Donaghy P, Thomas AJ, O’Brien JT. Amyloid PET imaging inLewy body disorders. Am J Geriatr Psychiatry 2013 Jul 3. pii:S1064-7481(13)00168-1. doi: 10.1016/j.jagp.2013.03.001.

40. Gomperts SN, Locascio JJ, Rentz D, et al. Amyloid is linked tocognitive decline in patients with Parkinson disease withoutdementia. Neurology 2013;80:85-91.

B U R N E T A L


info:doi/10.1016/j.jagp.2013.03.001

41. Defina LF, Willis BL, Radford NB, et al. The association betweenmidlife cardiorespiratory fitness levels and later-life dementia: acohort study. Ann Intern Med 2013;158:162-168.

42. Liang KY, Mintun MA, Fagan AM, et al. Exercise and Alzheimer’sdisease biomarkers in cognitively normal older adults. Ann Neurol2010;68:311-318.

43. Gons RA, Tuladhar AM, de Laat KF, et al. Physical activity isrelated to the structural integrity of cerebral white matter. Neurol-ogy 2013;81:971-976.

44. Head D, Bugg JM, Goate AM, et al. Exercise engagement as amoderator of the effects of APOE genotype on amyloid deposition.Arch Neurol 2012;69:636-643.

45. Hindle JV, Petrelli A, Clare L, Kalbe E. Nonpharmacologicalenhancement of cognitive function in Parkinson’s disease: a system-atic review. Mov Disord 2013;28:1034-1049.

46. Tanaka K, Quadros ACJ, Santos RF, Stella F, Gobbi LT, Gobbi S.Benefits of physical exercise on executive functions in older peoplewith Parkinson’s disease. Brain Cogn 2009;69:435-441.

47. Cruise KE, Bucks RS, Loftus AM, Newton RU, Pegoraro R,Thomas MG. Exercise and Parkinson’s: benefits for cognition andquality of life. Acta Neurol Scand 2011;123:13-19.

48. Petzinger GM, Fisher BE, McEwen S, Beeler JA, Walsh JP, JakowecMW. Exercise-enhanced neuroplasticity targeting motor and cognitivecircuitry in Parkinson’s disease. Lancet Neurol 2013;12:716-726.

49. Edwards JD, Hauser RA, O’Connor ML, Vald�es EG, ZesiewiczTA, Uc EY. Randomized trial of cognitive speed of processingtraining in Parkinson disease. Neurology 2013;81:1284-1290.

50. Freund HJ, Kuhn J, Lenartz D, et al. Cognitive functions in apatient with Parkinson-dementia syndrome undergoing deep brainstimulation. Arch Neurol 2009;66:781-785.

51. Hescham S, Lim LW, Jahanshahi A, Blokland A, Temel Y. Deepbrain stimulation in dementia-related disorders. Neurosci BiobehavRev 2013;37:2666-2675.

52. Laxton AW, Tang-Wai DF, McAndrews MP, et al. A phase I trialof deep brain stimulation of memory circuits in Alzheimer’s dis-ease. Ann Neurol 2010;68:521-534.

53. Gratwicke J, Kahan J, Zrinzo L, et al. The nucleus basalis of Mey-nert: a new target for deep brain stimulation in dementia? Neuro-sci Biobehav Rev 2013;37:2676-2688.

54. Boggio PS, Ferrucci R, Rigonatti SP, et al. Effects of transcranialdirect current stimulation on working memory in patients withParkinson’s disease. J Neurol Sci 2006;249:31-38.



cognition in movement disorders: where can we hope to be in ten years?

Documents